Twin Seal DBB Valve Pressure Relief

[atmengr]

Tried to search for this and talk with industry experts, but no conclusive answer so far....

A General Twin Seal valve, or a similar single body double block and bleed valve, often uses a device to relieve the pressure in the cavity between the 2 seals to the upstream (or downstream) side of the valve.

Is this device a "pressure relief device"? Does this device need to be inspected or replaced on an interval? Does anyone know of examples where this mechanism failed and caused overpressure of the valve? How would this overpressure most likely manifest?

I have several locations where I would like to install these valves to solve leak through problems in isolation, but I am wary that I will be solving one problem and creating another one that is not so obvious.

Thanks in advance!
Jacque'

 

[dcasto]

As operator, its your choice of what to call it. In 1993 we replaced actual springload 3/4" PSV's with a check valve that had a 25 psi differential spring in it. We tied the outlet to one side of the valve that had a thermal protection system on it or was a buried line not subject to thermal expansion. We worked with Foster valve to tap the flange of thier DBB (SAF T Seal knock off) valve and adopted that valve with the check valve attached as our standard. Our maintenance proceedures called for a an annual or semi annual inspection of the valves (grease, verify movement, corrosion inspection, and we pulled the check valve to verify it was not "stuck and still working". We did not cailbrate the spring pressure mechanically, but our inspectors knew about how much resistance the 25 psig spring had in it. The valve bodies were either ANSI 600, 900, or 1500. I think we may have set the 1500 to 100 psig differential.

Bottom line, we got out of the annual or semi annaul relief valve testing per DOT 195, 192. We also have no move venting when and if there was a thermal relief or a valve failure.

There rare cases of thermal overpressure on a DBB are very rare. I would even guess it would be nearly impossible, but you still need to have the system. The rests could be as catastophic as a valve body failure or as minor as loosing one of the seats.

 

[Sircrashalot]

Last year I went though an exercise to review the installation of PSV’s on some API 10,000 ball valves I supplied to an FPSO. These valves were installed on the gas injection system and had double piston effect seats, effectively the same as your twin seal valve in that if the upstream seat fails, the downstream seat can retain full line pressure. As per API 6D, the valve required a PSV to relieve cavity pressure in the event of a fire, and as the supplier I supplied the valves with the PSV’s installed.

During the client’s operational review prior to startup, they realized that they did not have any of these PSV’s piped to the flare system, and if they were to inadvertently lift, they would relieve direct to atmosphere and likely be picked up by gas detectors and shut the plant down. They were also concerned with having to do maintenance on these and issues with having to shutdown the lines to take the PSV’s out for re-calibration.

They considered the best option was to remove the PSV’s. They justified this in a couple ways:
1) As the valves were in gas service and the valves themselves were of a huge mass compared to the volume of gas in the cavity, the amount of heat energy required to cause an overpressure situation would have rendered the entire FPSO a twisted wreck on the ocean floor, the chances of a valve failure was very slim.
2) The soft seals (body seals) in the valve would be the weak point in the valve and would blow out prior to any catastrophic failure of any metal parts.

Still, I think usually the safest option is to provide for some form of cavity relief and to conduct maintenance on it. However, for the above I agreed with the operator that in the long run it was safer for them not to use the PSV’s. It really depends on the situation.

 

[dcasto]

The safest and most effective way is that mentioned in the general twin original post. This idea (of a spring loaded check valve) was pioneered with the Foster valve in the late 1980's for all the reasons Sircrashalot mentioned and then some (think benzene).

Relying on a seal to fail to protect you isn't always the right thing. What would happen a liquid was trapped in the valve body?